Tuberculosis (TB) remains a major health threat in Harlem where TB case rates markedly exceed national rates. A recent statement by the Advisory Council on the Elimination of TB highlights the importance of identifying persons with latent TB infection (LTBI), providing appropriate preventive treatment for those at high risk for progression to TB disease, and ensuring completion of treatment by at least 80% of those who initiate therapy. However, completion of LTBI treatment is challenging as patients are asymptomatic, are often unaware of the availability of effective preventive therapy, have unrealistic optimism do not appreciate the severity of TB disease and may overestimate the potential adverse events. In addition, they may have many competing priorities in their lives and few support networks. Finally, they may not have role models who have adopted this precautionary behavior. We propose to conduct a randomized clinical trial to compare the efficacy and cost-effectiveness of an experimental intervention when compared to current clinical practice. The experimental intervention will utilize key behavior models that are especially suited to LTBI and its treatment, the Health Belief Model (HBM) and the Precaution Adoption Process Model (PAPM). The HBM focuses on the patient's perception of susceptibility, severity, benefits, preservation of health, barriers, and self-efficacy. The PAPM will permit assessment of the patient's stage in adopting LTBI treatment and facilitate delivery of stage-specific intervention. Peers, who themselves have been treated for LTBI, will be utilized to deliver the experimental intervention as they are uniquely equipped to intervene on the perceptual level and to provide patients with much-needed role models. It is estimated that the sample size will be 360 participants. Participants will undergo baseline and regular follow assessments to determine adherence with LTBI treatment via several methods: self- report, computer touch-screen method, MEMs caps, pharmacy records and provider/peer assessments. In addition, the following data will be collected: demographic characteristics, PAPM behavior stage, social support, life stressors, quality of life, substance use, mental illness, literacy, perceived benefit/perceived barriers, self barriers, self-efficacy, substance use, TB knowledge and attitudes, and health care utilization. It is especially appropriate to conduct this study in the Harlem community where patients with LTBI are often members of those exact groups recently identified at high risk for progression to TB disease and who are currently targeted for LTBI treatment. Additionally, the recent availability of a 2-month short course LTBI treatment regimen offers an opportunity to assess adherence interventions for this condition with the goal of achieving high treatment completion rates. The identification of a practical, generalizable, efficacious as well as cost-effective intervention to enhance adherence with this pharmacologic therapeutic regimen as likely to have a substantial impact on TB control in this and other communities.